P
US10312793B2ActiveUtilityPatentIndex 38

Control of electrical converter based on optimized pulse patterns

Assignee: ABB SCHWEIZ AGPriority: Jul 9, 2015Filed: Jan 9, 2018Granted: Jun 4, 2019
Est. expiryJul 9, 2035(~9 yrs left)· nominal 20-yr term from priority
Inventors:ROHR EDUARDOAL-HOKAYEM PETERGEYER TOBIAS
H02M 7/53873H02M 1/12H02M 7/217H02M 2001/0012H02M 2001/0038H02M 1/0038H02M 1/0012
38
PatentIndex Score
0
Cited by
36
References
20
Claims

Abstract

A method for operating an electrical converter including: determining an optimized pulse pattern from a fundamental voltage reference for the electrical converter, wherein the optimized pulse pattern is determined from a first lookup table and includes discrete voltage amplitude values changing at predefined switching instants; determining a harmonic content reference from the fundamental voltage reference based on a second lookup table, wherein the harmonic content reference is a harmonic current reference determined from the frequency spectrum of a current of the electrical converter or the harmonic content reference is a filtered voltage reference determined by applying a first order frequency filter to a voltage, which current or voltage is generated, when the optimized pulse pattern is applied to the electrical converter; determining a harmonic content error from the harmonic content reference by subtracting an estimated output voltage and/or estimated output current of the electrical converter from the harmonic content reference; modifying the optimized pulse pattern by timeshifting switching instants such that the fundamental voltage reference is tracked and the harmonic content error is corrected by the timeshifted switching instants; applying the modified optimized pulse pattern to semiconductor switches of the electrical converter.

Claims

exact text as granted — not AI-modified
The invention claimed is: 
     
       1. A method for operating an electrical converter, the method comprising:
 determining an optimized pulse pattern ( v (t)) from a fundamental voltage reference ( v   ctr   1 (t)) for the electrical converter, wherein the optimized pulse pattern ( v (t)) is determined from a first lookup table and comprises discrete voltage amplitude values changing at predefined switching instants; 
 determining a harmonic content reference (ī h (t),  ψ (t)) from the fundamental voltage reference ( v   ctr   1 (t)) based on a second lookup table, wherein the harmonic content reference is a harmonic current reference (ī h (t)) determined from the frequency spectrum of a current of the electrical converter or the harmonic content reference is a filtered voltage reference ( ψ (t)) determined by applying a first order frequency filter to a voltage, which current or voltage is generated, when the optimized pulse pattern is applied to the electrical converter; 
 determining a harmonic content error (ĩ(t),  ψ (t)) from the harmonic content reference (ī h (t),  ψ (t)) by subtracting an estimated output voltage (ψ(t)) and/or estimated output current (i(t)) of the electrical converter from the harmonic content reference (ī h (t),  ψ (t)); 
 modifying the optimized pulse pattern ( v (t)) by timeshifting switching instants such that the fundamental voltage reference ( v   ctr   1 (t)) is tracked and the harmonic content error (ĩ(t),{tilde over (ψ)}(t)) is corrected by the timeshifted switching instants; 
 applying the modified optimized pulse pattern (v(t)) to semiconductor switches of the electrical converter. 
 
     
     
       2. The method of  claim 1 , wherein the harmonic content error (ĩ(t), {tilde over (ψ)}(t)) is corrected by minimizing a cost function, which comprises a term quadratic in the harmonic content error and/or a term quadratic in timeshifts of the switching instants. 
     
     
       3. The method of  claim 2 , wherein the timeshifts are applied in a time window ending at a prediction horizon; and/or
 wherein the cost function is based on the harmonic content error at the end of the prediction horizon. 
 
     
     
       4. The method of  claim 3 , wherein the cost function is minimized subject to the constraint that an order of switching instants does not change. 
     
     
       5. The method of  claim 3 , wherein the cost function is minimized by solving a quadratic programming. 
     
     
       6. The method of  claim 3 , further comprising:
 determining the fundamental voltage reference ( v   ctr   1 (t)) from a fundamental current reference (ī 1 (t)) and an estimated grid voltage (υ g ). 
 
     
     
       7. The method of  claim 3 ,
 wherein the harmonic content error ({tilde over (ψ)}(t)) is the difference of the filtered voltage reference ( ψ (t)) and an estimated filtered voltage (ψ(t)); 
 wherein the estimated filtered voltage (ψ(t)) is determined by online applying the first order frequency filter to the estimated output voltage of the electrical converter. 
 
     
     
       8. The method of  claim 2 , wherein the cost function is minimized subject to the constraint that an order of switching instants does not change. 
     
     
       9. The method of  claim 8 , wherein the cost function is minimized by solving a quadratic programming. 
     
     
       10. The method of  claim 2 , wherein the cost function is minimized by solving a quadratic programming. 
     
     
       11. The method of  claim 2 , further comprising:
 determining the fundamental voltage reference ( v   ctr   1 (t)) from a fundamental current reference (ī 1 (t)) and an estimated grid voltage (υ g ). 
 
     
     
       12. The method of  claim 2 ,
 wherein the harmonic content error ({tilde over (ψ)}(t)) is the difference of the filtered voltage reference ( ψ (t)) and an estimated filtered voltage (ψ(t)); 
 wherein the estimated filtered voltage (ψ(t)) is determined by online applying the first order frequency filter to the estimated output voltage of the electrical converter. 
 
     
     
       13. The method of  claim 1 , further comprising:
 determining the fundamental voltage reference ( v   ctr   1 (t)) from a fundamental current reference (ī 1 (t)) and an estimated grid voltage (υ g ). 
 
     
     
       14. The method of  claim 1 ,
 wherein the harmonic content error ({tilde over (ψ)}(t)) is the difference of the filtered voltage reference ( ψ (t)) and an estimated filtered voltage (ψ(t)); 
 wherein the estimated filtered voltage (ψ(t)) is determined by online applying the first order frequency filter to the estimated output voltage of the electrical converter. 
 
     
     
       15. The method of  claim 1 ,
 wherein the fundamental voltage reference ( v   ctr   1 (t)) is determined from a fundamental current reference (ī 1 (t)) and an error provided by an external controller. 
 
     
     
       16. The method of  claim 1 ,
 wherein the optimized pulse patterns ( v (t)) in the first lookup table have been calculated offline. 
 
     
     
       17. The method of  claim 1 ,
 wherein the harmonic content references (ī h (t),  ψ (t)) in the second lookup table have been calculated offline from the corresponding optimized pulse patterns applied to a model of the electrical converter. 
 
     
     
       18. A controller for an electrical converter, which controller is adapted for controlling the electrical converter, the controller operable to:
 determine an optimized pulse pattern ( v (t)) from a fundamental voltage reference ( v   ctr   1 (t)) for the electrical converter, wherein the optimized pulse pattern ( v (t)) is determined from a first lookup table and comprises discrete voltage amplitude values changing at predefined switching instants; 
 determine a harmonic content reference (ī h (t),  ψ (t)) from the fundamental voltage reference ( v   ctr   1 (t)) based on a second lookup table, wherein the harmonic content reference is a harmonic current reference (ī h (t)) determined from the frequency spectrum of a current of the electrical converter or the harmonic content reference is a filtered voltage reference ( ψ (t)) determined by applying a first order frequency filter to a voltage, which current or voltage is generated, when the optimized pulse pattern is applied to the electrical converter; 
 determine a harmonic content error (ĩ(t), {tilde over (ψ)}(t)) from the harmonic content reference (ī h (t),  ψ (t)) by subtracting an estimated output voltage (ψ(t)) and/or estimated output current (i(t)) of the electrical converter from the harmonic content reference (ī h ,  ψ (t)); 
 modify the optimized pulse pattern ( v (t)) by timeshifting switching instants such that the fundamental voltage reference ( v   ctr   1 (t)) is tracked and the harmonic content error (ĩ(t), {tilde over (ψ)}(t)) is corrected by the timeshifted switching instants; and 
 apply the modified optimized pulse pattern (v(t)) to semiconductor switches of the electrical converter. 
 
     
     
       19. A converter system, comprising:
 an electrical converter; and 
 a controller according to  claim 18 . 
 
     
     
       20. The converter system of  claim 19 ,
 wherein the converter is a voltage source converter for converting an AC input voltage into a DC output voltage.

Cited by (0)

No later patents cite this yet.

References (0)

No backward citations on record.